Marker clustering for performance
Interactive tracking and mapping
The project interested all of us because we have seen job scheduling algorithms in the real-world and wanted to take a stab at implementing them in the real world. We also wanted to stretch our knowledge of software engineering through learning to use various technologies such as MongoDB, Node.js, Twilio, React, Leaflet, and our brains ;)
What it does
Chevron WorkOrder is a full-stack service platform that tracks work orders that are submitted and the technicians that are completing them. Our platform optimizes assigned based on many factors, including knowing where technicians and facilities are, what they are certified to repair, how long they are planning to be there, other facility locations, as well as the work order priority.
In addition, we wanted to take it a step further and build out a full platform for Chevron that includes interactive SMS notifications for technicians to update their statuses, a real-time visualization of key business performance metrics, and a live-updating map to track worker and facility locations. We believe that these additional features, in companion with the original challenge specs, provide a superior analytics and work tracking experience for Chevron's users and technicians.
How we built it
We utilized Node.js with Express for the backend services, React for the frontend, and Twilio as a microservice for handling SMS flows. The optimization algorithm was implemented in Node.js using multivariate analysis and normalization.
Challenges we ran into
With so many features across the full stack that we wanted to accomplish, we had to learn many new technologies in order to implement them. We also ran into challenges with integrated all of our features into one consistent platform.
Accomplishments that we're proud of
We are proud that in less than 36 hours, we implemented more than what was required from the spec. Additionally, all our our data is live and none of the interactions with Chevron WorkOrder involve any hardcoded data.
What we learned
We learned best practices for working on and integrating features in parallel.
What's next for Chevron WorkOrders
We will continue to work on this and build out this full-fledged analytics platform.